Models
GitHub
Discord
Turbo
Sign in
Download
Models
Download
GitHub
Discord
Sign in
clore
/
qwen
:latest
1
Download
Updated
1 month ago
Cancel
qwen:latest
...
/
model
2e5c6a113918 · 142GB
Metadata
general.architecture
qwen3moe
qwen3moe
general.file_type
Q4_K_M
Q4_K_M
qwen3moe.attention.head_count
64
64
qwen3moe.attention.head_count_kv
4
4
qwen3moe.attention.key_length
128
128
qwen3moe.attention.layer_norm_rms_epsilon
1e-06
1e-06
qwen3moe.attention.value_length
128
128
qwen3moe.block_count
94
94
qwen3moe.context_length
40960
40960
qwen3moe.embedding_length
4096
4096
qwen3moe.expert_count
128
128
qwen3moe.expert_feed_forward_length
1536
1536
qwen3moe.expert_used_count
8
8
qwen3moe.feed_forward_length
12288
12288
qwen3moe.rope.freq_base
1e+06
1e+06
tokenizer.ggml.add_bos_token
false
false
tokenizer.ggml.eos_token_id
151645
151645
tokenizer.ggml.merges
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
[Ġ Ġ, ĠĠ ĠĠ, i n, Ġ t, ĠĠĠĠ ĠĠĠĠ, ...]
tokenizer.ggml.model
gpt2
gpt2
tokenizer.ggml.padding_token_id
151654
151654
tokenizer.ggml.pre
qwen2
qwen2
tokenizer.ggml.token_type
[1, 1, 1, 1, 1, ...]
[1, 1, 1, 1, 1, ...]
tokenizer.ggml.tokens
[!, ", #, $, %, ...]
[!, ", #, $, %, ...]
quantize.imatrix.chunks_count
685
685
quantize.imatrix.dataset
unsloth_calibration_Qwen3-235B-A22B.txt
unsloth_calibration_Qwen3-235B-A22B.txt
quantize.imatrix.entries_count
744
744
quantize.imatrix.file
Qwen3-235B-A22B-GGUF/imatrix_unsloth.dat
Qwen3-235B-A22B-GGUF/imatrix_unsloth.dat
split.count
3
3
split.no
0
0
split.tensors.count
1131
1131
Tensor
Name
Type
Shape
token_embd.weight
Q4_K
Q4_K
[4096, 151936]
blk.0
blk.0.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.0.attn_k_norm.weight
F32
F32
[128]
blk.0.attn_norm.weight
F32
F32
[4096]
blk.0.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.0.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.0.attn_q_norm.weight
F32
F32
[128]
blk.0.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.0.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.0.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.0.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.0.ffn_norm.weight
F32
F32
[4096]
blk.0.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.1
blk.1.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.1.attn_k_norm.weight
F32
F32
[128]
blk.1.attn_norm.weight
F32
F32
[4096]
blk.1.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.1.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.1.attn_q_norm.weight
F32
F32
[128]
blk.1.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.1.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.1.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.1.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.1.ffn_norm.weight
F32
F32
[4096]
blk.1.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.2
blk.2.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.2.attn_k_norm.weight
F32
F32
[128]
blk.2.attn_norm.weight
F32
F32
[4096]
blk.2.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.2.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.2.attn_q_norm.weight
F32
F32
[128]
blk.2.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.2.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.2.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.2.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.2.ffn_norm.weight
F32
F32
[4096]
blk.2.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.3
blk.3.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.3.attn_k_norm.weight
F32
F32
[128]
blk.3.attn_norm.weight
F32
F32
[4096]
blk.3.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.3.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.3.attn_q_norm.weight
F32
F32
[128]
blk.3.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.3.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.3.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.3.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.3.ffn_norm.weight
F32
F32
[4096]
blk.3.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.4
blk.4.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.4.attn_k_norm.weight
F32
F32
[128]
blk.4.attn_norm.weight
F32
F32
[4096]
blk.4.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.4.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.4.attn_q_norm.weight
F32
F32
[128]
blk.4.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.4.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.4.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.4.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.4.ffn_norm.weight
F32
F32
[4096]
blk.4.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.5
blk.5.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.5.attn_k_norm.weight
F32
F32
[128]
blk.5.attn_norm.weight
F32
F32
[4096]
blk.5.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.5.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.5.attn_q_norm.weight
F32
F32
[128]
blk.5.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.5.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.5.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.5.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.5.ffn_norm.weight
F32
F32
[4096]
blk.5.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.6
blk.6.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.6.attn_k_norm.weight
F32
F32
[128]
blk.6.attn_norm.weight
F32
F32
[4096]
blk.6.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.6.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.6.attn_q_norm.weight
F32
F32
[128]
blk.6.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.6.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.6.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.6.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.6.ffn_norm.weight
F32
F32
[4096]
blk.6.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.7
blk.7.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.7.attn_k_norm.weight
F32
F32
[128]
blk.7.attn_norm.weight
F32
F32
[4096]
blk.7.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.7.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.7.attn_q_norm.weight
F32
F32
[128]
blk.7.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.7.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.7.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.7.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.7.ffn_norm.weight
F32
F32
[4096]
blk.7.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.8
blk.8.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.8.attn_k_norm.weight
F32
F32
[128]
blk.8.attn_norm.weight
F32
F32
[4096]
blk.8.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.8.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.8.attn_q_norm.weight
F32
F32
[128]
blk.8.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.8.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.8.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.8.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.8.ffn_norm.weight
F32
F32
[4096]
blk.8.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.9
blk.9.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.9.attn_k_norm.weight
F32
F32
[128]
blk.9.attn_norm.weight
F32
F32
[4096]
blk.9.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.9.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.9.attn_q_norm.weight
F32
F32
[128]
blk.9.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.9.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.9.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.9.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.9.ffn_norm.weight
F32
F32
[4096]
blk.9.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.10
blk.10.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.10.attn_k_norm.weight
F32
F32
[128]
blk.10.attn_norm.weight
F32
F32
[4096]
blk.10.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.10.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.10.attn_q_norm.weight
F32
F32
[128]
blk.10.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.10.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.10.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.10.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.10.ffn_norm.weight
F32
F32
[4096]
blk.10.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.11
blk.11.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.11.attn_k_norm.weight
F32
F32
[128]
blk.11.attn_norm.weight
F32
F32
[4096]
blk.11.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.11.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.11.attn_q_norm.weight
F32
F32
[128]
blk.11.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.11.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.11.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.11.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.11.ffn_norm.weight
F32
F32
[4096]
blk.11.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.12
blk.12.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.12.attn_k_norm.weight
F32
F32
[128]
blk.12.attn_norm.weight
F32
F32
[4096]
blk.12.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.12.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.12.attn_q_norm.weight
F32
F32
[128]
blk.12.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.12.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.12.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.12.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.12.ffn_norm.weight
F32
F32
[4096]
blk.12.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.13
blk.13.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.13.attn_k_norm.weight
F32
F32
[128]
blk.13.attn_norm.weight
F32
F32
[4096]
blk.13.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.13.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.13.attn_q_norm.weight
F32
F32
[128]
blk.13.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.13.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.13.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.13.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.13.ffn_norm.weight
F32
F32
[4096]
blk.13.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.14
blk.14.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.14.attn_k_norm.weight
F32
F32
[128]
blk.14.attn_norm.weight
F32
F32
[4096]
blk.14.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.14.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.14.attn_q_norm.weight
F32
F32
[128]
blk.14.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.14.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.14.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.14.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.14.ffn_norm.weight
F32
F32
[4096]
blk.14.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.15
blk.15.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.15.attn_k_norm.weight
F32
F32
[128]
blk.15.attn_norm.weight
F32
F32
[4096]
blk.15.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.15.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.15.attn_q_norm.weight
F32
F32
[128]
blk.15.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.15.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.15.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.15.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.15.ffn_norm.weight
F32
F32
[4096]
blk.15.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.16
blk.16.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.16.attn_k_norm.weight
F32
F32
[128]
blk.16.attn_norm.weight
F32
F32
[4096]
blk.16.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.16.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.16.attn_q_norm.weight
F32
F32
[128]
blk.16.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.16.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.16.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.16.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.16.ffn_norm.weight
F32
F32
[4096]
blk.16.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.17
blk.17.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.17.attn_k_norm.weight
F32
F32
[128]
blk.17.attn_norm.weight
F32
F32
[4096]
blk.17.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.17.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.17.attn_q_norm.weight
F32
F32
[128]
blk.17.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.17.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.17.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.17.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.17.ffn_norm.weight
F32
F32
[4096]
blk.17.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.18
blk.18.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.18.attn_k_norm.weight
F32
F32
[128]
blk.18.attn_norm.weight
F32
F32
[4096]
blk.18.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.18.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.18.attn_q_norm.weight
F32
F32
[128]
blk.18.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.18.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.18.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.18.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.18.ffn_norm.weight
F32
F32
[4096]
blk.18.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.19
blk.19.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.19.attn_k_norm.weight
F32
F32
[128]
blk.19.attn_norm.weight
F32
F32
[4096]
blk.19.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.19.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.19.attn_q_norm.weight
F32
F32
[128]
blk.19.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.19.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.19.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.19.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.19.ffn_norm.weight
F32
F32
[4096]
blk.19.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.20
blk.20.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.20.attn_k_norm.weight
F32
F32
[128]
blk.20.attn_norm.weight
F32
F32
[4096]
blk.20.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.20.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.20.attn_q_norm.weight
F32
F32
[128]
blk.20.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.20.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.20.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.20.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.20.ffn_norm.weight
F32
F32
[4096]
blk.20.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.21
blk.21.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.21.attn_k_norm.weight
F32
F32
[128]
blk.21.attn_norm.weight
F32
F32
[4096]
blk.21.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.21.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.21.attn_q_norm.weight
F32
F32
[128]
blk.21.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.21.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.21.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.21.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.21.ffn_norm.weight
F32
F32
[4096]
blk.21.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.22
blk.22.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.22.attn_k_norm.weight
F32
F32
[128]
blk.22.attn_norm.weight
F32
F32
[4096]
blk.22.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.22.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.22.attn_q_norm.weight
F32
F32
[128]
blk.22.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.22.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.22.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.22.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.22.ffn_norm.weight
F32
F32
[4096]
blk.22.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.23
blk.23.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.23.attn_k_norm.weight
F32
F32
[128]
blk.23.attn_norm.weight
F32
F32
[4096]
blk.23.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.23.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.23.attn_q_norm.weight
F32
F32
[128]
blk.23.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.23.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.23.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.23.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.23.ffn_norm.weight
F32
F32
[4096]
blk.23.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.24
blk.24.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.24.attn_k_norm.weight
F32
F32
[128]
blk.24.attn_norm.weight
F32
F32
[4096]
blk.24.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.24.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.24.attn_q_norm.weight
F32
F32
[128]
blk.24.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.24.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.24.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.24.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.24.ffn_norm.weight
F32
F32
[4096]
blk.24.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.25
blk.25.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.25.attn_k_norm.weight
F32
F32
[128]
blk.25.attn_norm.weight
F32
F32
[4096]
blk.25.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.25.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.25.attn_q_norm.weight
F32
F32
[128]
blk.25.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.25.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.25.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.25.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.25.ffn_norm.weight
F32
F32
[4096]
blk.25.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.26
blk.26.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.26.attn_k_norm.weight
F32
F32
[128]
blk.26.attn_norm.weight
F32
F32
[4096]
blk.26.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.26.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.26.attn_q_norm.weight
F32
F32
[128]
blk.26.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.26.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.26.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.26.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.26.ffn_norm.weight
F32
F32
[4096]
blk.26.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.27
blk.27.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.27.attn_k_norm.weight
F32
F32
[128]
blk.27.attn_norm.weight
F32
F32
[4096]
blk.27.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.27.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.27.attn_q_norm.weight
F32
F32
[128]
blk.27.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.27.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.27.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.27.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.27.ffn_norm.weight
F32
F32
[4096]
blk.27.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.28
blk.28.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.28.attn_k_norm.weight
F32
F32
[128]
blk.28.attn_norm.weight
F32
F32
[4096]
blk.28.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.28.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.28.attn_q_norm.weight
F32
F32
[128]
blk.28.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.28.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.28.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.28.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.28.ffn_norm.weight
F32
F32
[4096]
blk.28.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.29
blk.29.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.29.attn_k_norm.weight
F32
F32
[128]
blk.29.attn_norm.weight
F32
F32
[4096]
blk.29.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.29.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.29.attn_q_norm.weight
F32
F32
[128]
blk.29.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.29.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.29.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.29.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.29.ffn_norm.weight
F32
F32
[4096]
blk.29.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.30
blk.30.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.30.attn_k_norm.weight
F32
F32
[128]
blk.30.attn_norm.weight
F32
F32
[4096]
blk.30.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.30.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.30.attn_q_norm.weight
F32
F32
[128]
blk.30.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.30.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
blk.30.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.30.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.30.ffn_norm.weight
F32
F32
[4096]
blk.30.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.31
blk.31.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.31.attn_k_norm.weight
F32
F32
[128]
blk.31.attn_norm.weight
F32
F32
[4096]
blk.31.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.31.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.31.attn_q_norm.weight
F32
F32
[128]
blk.31.attn_v.weight
Q6_K
Q6_K
[4096, 512]
blk.31.ffn_down_exps.weight
Q6_K
Q6_K
[1536, 4096, 128]
blk.31.ffn_gate_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.31.ffn_gate_inp.weight
F32
F32
[4096, 128]
blk.31.ffn_norm.weight
F32
F32
[4096]
blk.31.ffn_up_exps.weight
Q4_K
Q4_K
[4096, 1536, 128]
blk.32
blk.32.attn_k.weight
Q4_K
Q4_K
[4096, 512]
blk.32.attn_k_norm.weight
F32
F32
[128]
blk.32.attn_norm.weight
F32
F32
[4096]
blk.32.attn_output.weight
Q4_K
Q4_K
[8192, 4096]
blk.32.attn_q.weight
Q4_K
Q4_K
[4096, 8192]
blk.32.attn_q_norm.weight
F32
F32
[128]
blk.32.attn_v.weight
Q4_K
Q4_K
[4096, 512]
blk.32.ffn_down_exps.weight
Q4_K
Q4_K
[1536, 4096, 128]
output.weight
Q6_K
Q6_K
[4096, 151936]
output_norm.weight
F32
F32
[4096]